Error Span Annotation: A Balanced Approach for Human Evaluation of Machine Translation

Tom Kocmi, Vilém Zouhar, Eleftherios Avramidis, Roman Grundkiewicz, Marzena Karpinska +3 more
6/17/2024
cs.CL

No official code found. Best implementation: wmt-conference/ErrorSpanAnnotation (9 stars, GPL-3.0, last commit 2025-07-25).

Abstract

High-quality Machine Translation (MT) evaluation relies heavily on human judgments. Comprehensive error classification methods, such as Multidimensional Quality Metrics (MQM), are expensive as they are time-consuming and can only be done by experts, whose availability may be limited especially for low-resource languages. On the other hand, just assigning overall scores, like Direct Assessment (DA), is simpler and faster and can be done by translators of any level, but is less reliable. In this paper, we introduce Error Span Annotation (ESA), a human evaluation protocol which combines the continuous rating of DA with the high-level error severity span marking of MQM. We validate ESA by comparing it to MQM and DA for 12 MT systems and one human reference translation (English to German) from WMT23. The results show that ESA offers faster and cheaper annotations than MQM at the same quality level, without the requirement of expensive MQM experts.

View on arXivView PDF

Code Implementations(1)

wmt-conference/ErrorSpanAnnotationCommunity100%
GPL-3.0
91CSS, HTMLMar 19, 20241 years agoGPL-3.0

Similar Papers

AI-Assisted Human Evaluation of Machine Translation

89% match1 repo92024

The Devil is in the Errors: Leveraging Large Language Models for Fine-grained Machine Translation Evaluation

88% match2023

MQM-APE: Toward High-Quality Error Annotation Predictors with Automatic Post-Editing in LLM Translation Evaluators

87% match1 repo152024

Learning from others' mistakes: Finetuning machine translation models with span-level error annotations

85% match2024

Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress

85% match1 repo32025

TranslationCorrect: A Unified Framework for Machine Translation Post-Editing with Predictive Error Assistance

85% match2025

Cite this paper

@article{kocmi2024error,
  title  = {Error Span Annotation: A Balanced Approach for Human Evaluation of Machine Translation},
  author = {Tom Kocmi and Vilém Zouhar and Eleftherios Avramidis and Roman Grundkiewicz and Marzena Karpinska and Maja Popović and Mrinmaya Sachan and Mariya Shmatova},
  year   = {2024},
  eprint = {2406.11580},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2406.11580v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2406.11580.svg)](https://codeofpaper.com/paper/2406.11580)

Discussion